diff --git a/mindspore/ccsrc/backend/kernel_compiler/cpu/assign_cpu_kernel.cc b/mindspore/ccsrc/backend/kernel_compiler/cpu/assign_cpu_kernel.cc
index a1c4cd115bf..4064190e192 100644
--- a/mindspore/ccsrc/backend/kernel_compiler/cpu/assign_cpu_kernel.cc
+++ b/mindspore/ccsrc/backend/kernel_compiler/cpu/assign_cpu_kernel.cc
@@ -53,9 +53,9 @@ bool AssignCPUKernel::Launch(const std::vector<AddressPtr> &inputs, const std::v
     MS_LOG(EXCEPTION) << "Memcpy size must <= max_size, but got memcpy size is : " << total_size
                       << ", max size is : " << max_size;
   }
-  int ret = memcpy_s(inputs[0]->addr, total_size, inputs[1]->addr, total_size);
+  int ret = memcpy_s(inputs[0]->addr, max_size, inputs[1]->addr, total_size);
   if (ret != 0) {
-    MS_LOG(EXCEPTION) << "memcpy_s error, errorno" << ret;
+    MS_LOG(EXCEPTION) << "memcpy_s error, error no " << ret;
   }
   return true;
 }
diff --git a/mindspore/ccsrc/backend/kernel_compiler/cpu/maximum_cpu_kernel.cc b/mindspore/ccsrc/backend/kernel_compiler/cpu/maximum_cpu_kernel.cc
index 5fa83d8487c..8e4ffd5dcd5 100644
--- a/mindspore/ccsrc/backend/kernel_compiler/cpu/maximum_cpu_kernel.cc
+++ b/mindspore/ccsrc/backend/kernel_compiler/cpu/maximum_cpu_kernel.cc
@@ -19,7 +19,6 @@
 
 namespace mindspore {
 namespace kernel {
-
 template <typename T>
 void MaximumCPUKernel<T>::InitKernel(const CNodePtr &kernel_node) {
   CheckParam(kernel_node);
@@ -216,6 +215,5 @@ void MaximumCPUKernel<T>::BroadcastArithTensors(const T *input_x, const T *input
     output[i] = MaximumFunc(input_x[i], input_y[i]);
   }
 }
-
 }  // namespace kernel
 }  // namespace mindspore
diff --git a/mindspore/nn/layer/__init__.py b/mindspore/nn/layer/__init__.py
index 4aecf85d976..3c4a601c1cc 100644
--- a/mindspore/nn/layer/__init__.py
+++ b/mindspore/nn/layer/__init__.py
@@ -17,7 +17,8 @@ Layer.
 
 The high-level components(Cells) used to construct the neural network.
 """
-from . import activation, normalization, container, conv, lstm, basic, embedding, pooling, image, quant, math, combined
+from . import activation, normalization, container, conv, lstm, basic, embedding, pooling, image, quant, math, \
+    combined, timedistributed
 from .activation import *
 from .normalization import *
 from .container import *
@@ -30,6 +31,7 @@ from .image import *
 from .quant import *
 from .math import *
 from .combined import *
+from .timedistributed import *
 
 __all__ = []
 __all__.extend(activation.__all__)
@@ -44,3 +46,4 @@ __all__.extend(image.__all__)
 __all__.extend(quant.__all__)
 __all__.extend(math.__all__)
 __all__.extend(combined.__all__)
+__all__.extend(timedistributed.__all__)
diff --git a/mindspore/nn/layer/timedistributed.py b/mindspore/nn/layer/timedistributed.py
new file mode 100644
index 00000000000..e7ee93b05bf
--- /dev/null
+++ b/mindspore/nn/layer/timedistributed.py
@@ -0,0 +1,138 @@
+# Copyright 2020 Huawei Technologies Co., Ltd
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# ============================================================================
+"""Time Distributed."""
+
+from mindspore.ops.primitive import constexpr, Primitive
+from mindspore.ops import Reshape, Transpose, Pack, Unpack
+from mindspore.common.dtype import tensor
+from ..cell import Cell
+
+__all__ = ['TimeDistributed']
+
+
+@constexpr
+def _check_reshape_pos(reshape_pos, inputs_shape, outputs_shape):
+    if reshape_pos >= len(outputs_shape) or inputs_shape[reshape_pos] != outputs_shape[reshape_pos]:
+        raise ValueError("The parameter reshape_with_axis is invalid in the input and output of TimeDistributed. "
+                         "You may try pass parameters without reshape_with_axis.")
+
+
+@constexpr
+def _check_expand_dims_axis(time_axis, ndim):
+    if time_axis > ndim:
+        raise ValueError("The parameter time_axis is invalid in the input. "
+                         "The value of time_axis should be in range of [{}, {}].".format(-ndim - 1, ndim))
+
+
+@constexpr
+def _generate_perm(axis_a, axis_b, length):
+    perm = tuple(range(length))
+    axis_a, axis_b = (axis_a, axis_b) if axis_a < axis_b else (axis_b, axis_a)
+    return perm[:axis_a] + perm[axis_a + 1: axis_b + 1] + (perm[axis_a],) + perm[axis_b + 1:]
+
+
+@constexpr
+def _check_data(flag):
+    if not flag:
+        raise TypeError("The inputs and outputs shuould be a Tensor.")
+
+
+@constexpr
+def _check_inputs_dim(shape):
+    if len(shape) < 3:
+        raise ValueError("The inputs should be at least 3D.")
+
+
+class TimeDistributed(Cell):
+    r"""
+    The time distributed layer.
+
+    Time distributed is a wrapper which allows to apply a layer to every temporal slice of an input.
+    And the input should be at least 3D.
+    There are two cases in the implementation.
+    When reshape_with_axis provided, the reshape method will be chosen, which is more efficient;
+    otherwise, the method of dividing the inputs along time axis will be used, which is more general.
+    For example, reshape_with_axis could not be provided when deal with batch normal.
+
+    Args:
+        layer(Union[Cell, Primitive]): The Cell or Primitive which will be wrapped.
+        time_axis(int): The axis of time_step.
+        reshape_with_axis(int): The axis which time_axis will be reshaped with. Default: 'None'.
+
+    Raises:
+        TypeError: If cell is not a Cell or Primitive.
+
+    inputs:
+        -**input**(Tensor)-Tensor of shape: math:'(N, T, *)'
+
+    Outputs:
+        Tensor of shape: math:'(N, T, *)'
+
+    Supported Platforms:
+        ``Ascend`` ``GPU`` ``CPU``
+
+    Examples:
+        >>> input = Tensor(np.random.random([32, 10, 3]), mindspore.float32)
+        >>> dense = nn.Dense(3, 6)
+        >>> net = TimeDistributed(dense, time_axis=1, reshape_with_axis=0)
+        >>> output = net(input)
+        >>> print(output.shape)
+        (32, 10, 6)
+    """
+
+    def __init__(self, layer, time_axis, reshape_with_axis=None):
+        if not isinstance(layer, (Cell, Primitive)):
+            raise TypeError("Please initialize TimeDistributed with mindspore.nn.Cell or "
+                            "mindspore.ops.Primitive instance. You passed: {input}".format(input=layer))
+        super(TimeDistributed, self).__init__()
+        self.layer = layer
+        self.time_axis = time_axis
+        self.reshape_with_axis = reshape_with_axis
+        self.transpose = Transpose()
+        self.reshape = Reshape()
+
+    def construct(self, inputs):
+        _check_data(isinstance(inputs, tensor))
+        _check_inputs_dim(inputs.shape)
+        time_axis = self.time_axis % len(inputs.shape)
+        if self.reshape_with_axis is not None:
+            reshape_with_axis = self.reshape_with_axis % len(inputs.shape)
+            inputs_shape = inputs.shape
+            time_axis_new = len(inputs_shape) - 2 if reshape_with_axis == len(inputs_shape) - 1 \
+                else (reshape_with_axis + 1 if time_axis > reshape_with_axis else
+                      reshape_with_axis - 1)
+            reshape_pos = time_axis_new if time_axis_new < reshape_with_axis else reshape_with_axis
+            perm = _generate_perm(time_axis_new, time_axis, len(inputs_shape))
+            inputs = self.transpose(inputs, perm)
+            inputs_shape_new = inputs.shape
+            inputs = self.reshape(inputs, inputs_shape_new[: reshape_pos] + (-1,) + inputs_shape_new[reshape_pos + 2:])
+            outputs = self.layer(inputs)
+            _check_data(isinstance(outputs, tensor))
+            _check_reshape_pos(reshape_pos, inputs.shape, outputs.shape)
+            outputs_shape_new = outputs.shape[:reshape_pos] + inputs_shape_new[reshape_pos: reshape_pos + 2]
+            if reshape_pos + 1 < len(outputs.shape):
+                outputs_shape_new += outputs.shape[reshape_pos + 1:]
+            return self.reshape(outputs, outputs_shape_new)
+
+        unpack = Unpack(time_axis)
+        inputs = unpack(inputs)
+        y = ()
+        for item in inputs:
+            outputs = self.layer(item)
+            _check_data(isinstance(outputs, tensor))
+            _check_expand_dims_axis(time_axis, outputs.ndim)
+            y += (outputs,)
+        y = Pack(time_axis)(y)
+        return y
diff --git a/tests/st/ops/cpu/test_time_distributed_op.py b/tests/st/ops/cpu/test_time_distributed_op.py
new file mode 100644
index 00000000000..f6d0a3641b8
--- /dev/null
+++ b/tests/st/ops/cpu/test_time_distributed_op.py
@@ -0,0 +1,198 @@
+# Copyright 2020 Huawei Technologies Co., Ltd
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# ============================================================================
+import numpy as np
+import pytest
+
+import mindspore
+import mindspore.context as context
+import mindspore.nn as nn
+import mindspore.ops as ops
+from mindspore import Tensor
+
+context.set_context(mode=context.GRAPH_MODE, device_target='CPU')
+
+
+class TestTimeDistributed(nn.Cell):
+    def __init__(self, cell, time_axis, reshape_with_axis=None):
+        super(TestTimeDistributed, self).__init__()
+        self.time_distributed = nn.TimeDistributed(cell, time_axis, reshape_with_axis)
+
+    def construct(self, inputs):
+        return self.time_distributed(inputs)
+
+
+@pytest.mark.level0
+@pytest.mark.platform_x86_cpu
+@pytest.mark.env_onecard
+def test_time_distributed_conv2d():
+    inputs = np.random.randint(0, 10, [32, 12, 10, 10])
+    conv2d = nn.Conv2d(12, 24, 4, has_bias=False, weight_init='normal')
+    output_expect = conv2d(Tensor(inputs, mindspore.float32)).asnumpy()
+    inputs = inputs.reshape([32, 1, 12, 10, 10]).repeat(6, axis=1)
+    time_distributed = TestTimeDistributed(conv2d, time_axis=1, reshape_with_axis=0)
+    output = time_distributed(Tensor(inputs, mindspore.float32)).asnumpy()
+    for i in range(output.shape[1]):
+        assert np.all(np.abs(output[:, i, :] - output_expect) < 1e-5)
+    print("Conv2D layer wrapped successful")
+
+
+@pytest.mark.level0
+@pytest.mark.platform_x86_cpu
+@pytest.mark.env_onecard
+def test_time_distributed_maxpool2d():
+    inputs = np.random.randint(0, 10, [32, 12, 10, 10])
+    pool = nn.MaxPool2d(kernel_size=3, stride=1)
+    output_expect = pool(Tensor(inputs, mindspore.float32)).asnumpy()
+    inputs = inputs.reshape([32, 1, 12, 10, 10]).repeat(6, axis=1)
+    time_distributed = TestTimeDistributed(pool, time_axis=1, reshape_with_axis=0)
+    output = time_distributed(Tensor(inputs, mindspore.float32)).asnumpy()
+    for i in range(output.shape[1]):
+        assert np.all(output[:, i, :] == output_expect)
+    print("MaxPooling2D layer wrapped successful")
+
+
+@pytest.mark.level0
+@pytest.mark.platform_x86_cpu
+@pytest.mark.env_onecard
+def test_time_distributed_dense():
+    inputs = np.random.randint(0, 10, [32, 10])
+    dense = nn.Dense(10, 6)
+    output_expect = dense(Tensor(inputs, mindspore.float32)).asnumpy()
+    inputs = inputs.reshape([32, 1, 10]).repeat(6, axis=1)
+    time_distributed = TestTimeDistributed(dense, time_axis=1, reshape_with_axis=0)
+    output = time_distributed(Tensor(inputs, mindspore.float32)).asnumpy()
+    for i in range(output.shape[1]):
+        assert np.all(output[:, i, :] == output_expect)
+    print("Dense layer wrapped successful")
+
+
+@pytest.mark.level0
+@pytest.mark.platform_x86_cpu
+@pytest.mark.env_onecard
+def test_time_distributed_dense_with_reshape_axis_not_first():
+    inputs = np.random.randint(0, 10, [32, 10])
+    dense = nn.Dense(10, 6)
+    output_expect = dense(Tensor(inputs, mindspore.float32)).asnumpy()
+    inputs = inputs.reshape([1, 32, 10]).repeat(6, axis=0)
+    time_distributed = TestTimeDistributed(dense, time_axis=0, reshape_with_axis=1)
+    output = time_distributed(Tensor(inputs, mindspore.float32)).asnumpy()
+    for i in range(output.shape[0]):
+        assert np.all(output[i, :] == output_expect)
+    print("Dense layer wrapped successful")
+
+
+@pytest.mark.level0
+@pytest.mark.platform_x86_cpu
+@pytest.mark.env_onecard
+def test_time_distributed_argmax():
+    inputs = np.random.randint(0, 10, [3, 4])
+    argmax = ops.Argmax(output_type=mindspore.int32, axis=1)
+    output_expect = argmax(Tensor(inputs, mindspore.float32)).asnumpy()
+    inputs = inputs.reshape([3, 1, 4]).repeat(6, axis=1)
+    time_distributed = TestTimeDistributed(argmax, time_axis=1, reshape_with_axis=0)
+    output = time_distributed(Tensor(inputs, mindspore.float32)).asnumpy()
+    for i in range(output.shape[1]):
+        assert np.all(output[:, i] == output_expect)
+    print("Argmax op wrapped successful")
+
+
+@pytest.mark.level0
+@pytest.mark.platform_x86_cpu
+@pytest.mark.env_onecard
+def test_time_distributed_flatten():
+    inputs = np.random.randint(0, 10, [3, 4, 5])
+    flatten = nn.Flatten()
+    output_expect = flatten(Tensor(inputs, mindspore.float32)).asnumpy()
+    inputs = inputs.reshape([3, 1, 4, 5]).repeat(6, axis=1)
+    time_distributed = TestTimeDistributed(flatten, time_axis=1, reshape_with_axis=0)
+    output = time_distributed(Tensor(inputs, mindspore.float32)).asnumpy()
+    for i in range(output.shape[1]):
+        assert np.all(output[:, i, :] == output_expect)
+    print("Flatten op wrapped successful")
+
+
+@pytest.mark.level0
+@pytest.mark.platform_x86_cpu
+@pytest.mark.env_onecard
+def test_time_distributed_conv2d_no_reshape_axis():
+    inputs = np.random.randint(0, 10, [32, 12, 10, 10])
+    conv2d = nn.Conv2d(12, 24, 4, has_bias=False, weight_init='normal')
+    output_expect = conv2d(Tensor(inputs, mindspore.float32)).asnumpy()
+    inputs = inputs.reshape([32, 1, 12, 10, 10]).repeat(6, axis=1)
+    time_distributed = TestTimeDistributed(conv2d, time_axis=1)
+    output = time_distributed(Tensor(inputs, mindspore.float32)).asnumpy()
+    for i in range(output.shape[1]):
+        assert np.all(output[:, i, :] == output_expect)
+    print("Conv2D layer with no reshape axis wrapped successful")
+
+
+@pytest.mark.level0
+@pytest.mark.platform_x86_cpu
+@pytest.mark.env_onecard
+def test_time_distributed_maxpool2d_no_reshape_axis():
+    inputs = np.random.randint(0, 10, [32, 12, 10, 10])
+    pool = nn.MaxPool2d(kernel_size=3, stride=1)
+    output_expect = pool(Tensor(inputs, mindspore.float32)).asnumpy()
+    inputs = inputs.reshape([32, 1, 12, 10, 10]).repeat(6, axis=1)
+    time_distributed = TestTimeDistributed(pool, time_axis=1)
+    output = time_distributed(Tensor(inputs, mindspore.float32)).asnumpy()
+    for i in range(output.shape[1]):
+        assert np.all(output[:, i, :] == output_expect)
+    print("MaxPooling2D layer with no reshape axis wrapped successful")
+
+
+@pytest.mark.level0
+@pytest.mark.platform_x86_cpu
+@pytest.mark.env_onecard
+def test_time_distributed_dense_no_reshape_axis():
+    inputs = np.random.randint(0, 10, [32, 10])
+    dense = nn.Dense(10, 6)
+    output_expect = dense(Tensor(inputs, mindspore.float32)).asnumpy()
+    inputs = inputs.reshape([32, 1, 10]).repeat(6, axis=1)
+    time_distributed = TestTimeDistributed(dense, time_axis=1)
+    output = time_distributed(Tensor(inputs, mindspore.float32)).asnumpy()
+    for i in range(output.shape[1]):
+        assert np.all(output[:, i, :] == output_expect)
+    print("Dense layer with no reshape axis wrapped successful")
+
+
+@pytest.mark.level0
+@pytest.mark.platform_x86_cpu
+@pytest.mark.env_onecard
+def test_time_distributed_argmax_no_reshape_axis():
+    inputs = np.random.randint(0, 10, [3, 4])
+    argmax = ops.Argmax(output_type=mindspore.int32, axis=1)
+    output_expect = argmax(Tensor(inputs, mindspore.float32)).asnumpy()
+    inputs = inputs.reshape([3, 1, 4]).repeat(6, axis=1)
+    time_distributed = TestTimeDistributed(argmax, time_axis=1)
+    output = time_distributed(Tensor(inputs, mindspore.float32)).asnumpy()
+    for i in range(output.shape[1]):
+        assert np.all(output[:, i] == output_expect)
+    print("Argmax op with no reshape axis wrapped successful")
+
+
+@pytest.mark.level0
+@pytest.mark.platform_x86_cpu
+@pytest.mark.env_onecard
+def test_time_distributed_flatten_no_reshape_axis():
+    inputs = np.random.randint(0, 10, [3, 4, 5])
+    flatten = nn.Flatten()
+    output_expect = flatten(Tensor(inputs, mindspore.float32)).asnumpy()
+    inputs = inputs.reshape([3, 1, 4, 5]).repeat(6, axis=1)
+    time_distributed = TestTimeDistributed(flatten, time_axis=1)
+    output = time_distributed(Tensor(inputs, mindspore.float32)).asnumpy()
+    for i in range(output.shape[1]):
+        assert np.all(output[:, i, :] == output_expect)
+    print("Flatten op with no reshape axis wrapped successful")
diff --git a/tests/st/ops/gpu/test_time_distributed_op.py b/tests/st/ops/gpu/test_time_distributed_op.py
new file mode 100644
index 00000000000..6ff85747b2a
--- /dev/null
+++ b/tests/st/ops/gpu/test_time_distributed_op.py
@@ -0,0 +1,198 @@
+# Copyright 2020 Huawei Technologies Co., Ltd
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# ============================================================================
+import numpy as np
+import pytest
+
+import mindspore
+import mindspore.context as context
+import mindspore.nn as nn
+import mindspore.ops as ops
+from mindspore import Tensor
+
+context.set_context(mode=context.GRAPH_MODE, device_target='GPU')
+
+
+class TestTimeDistributed(nn.Cell):
+    def __init__(self, cell, time_axis, reshape_with_axis=None):
+        super(TestTimeDistributed, self).__init__()
+        self.time_distributed = nn.TimeDistributed(cell, time_axis, reshape_with_axis)
+
+    def construct(self, inputs):
+        return self.time_distributed(inputs)
+
+
+@pytest.mark.level0
+@pytest.mark.platform_x86_gpu_training
+@pytest.mark.env_onecard
+def test_time_distributed_conv2d():
+    inputs = np.random.randint(0, 10, [32, 12, 10, 10])
+    conv2d = nn.Conv2d(12, 24, 4, has_bias=False, weight_init='normal')
+    output_expect = conv2d(Tensor(inputs, mindspore.float32)).asnumpy()
+    inputs = inputs.reshape([32, 1, 12, 10, 10]).repeat(6, axis=1)
+    time_distributed = TestTimeDistributed(conv2d, time_axis=1, reshape_with_axis=0)
+    output = time_distributed(Tensor(inputs, mindspore.float32)).asnumpy()
+    for i in range(output.shape[1]):
+        assert np.all(np.abs(output[:, i, :] - output_expect) < 1e-5)
+    print("Conv2D layer wrapped successful")
+
+
+@pytest.mark.level0
+@pytest.mark.platform_x86_gpu_training
+@pytest.mark.env_onecard
+def test_time_distributed_maxpool2d():
+    inputs = np.random.randint(0, 10, [32, 12, 10, 10])
+    pool = nn.MaxPool2d(kernel_size=3, stride=1)
+    output_expect = pool(Tensor(inputs, mindspore.float32)).asnumpy()
+    inputs = inputs.reshape([32, 1, 12, 10, 10]).repeat(6, axis=1)
+    time_distributed = TestTimeDistributed(pool, time_axis=1, reshape_with_axis=0)
+    output = time_distributed(Tensor(inputs, mindspore.float32)).asnumpy()
+    for i in range(output.shape[1]):
+        assert np.all(output[:, i, :] == output_expect)
+    print("MaxPooling2D layer wrapped successful")
+
+
+@pytest.mark.level0
+@pytest.mark.platform_x86_gpu_training
+@pytest.mark.env_onecard
+def test_time_distributed_dense():
+    inputs = np.random.randint(0, 10, [32, 10])
+    dense = nn.Dense(10, 6)
+    output_expect = dense(Tensor(inputs, mindspore.float32)).asnumpy()
+    inputs = inputs.reshape([32, 1, 10]).repeat(6, axis=1)
+    time_distributed = TestTimeDistributed(dense, time_axis=1, reshape_with_axis=0)
+    output = time_distributed(Tensor(inputs, mindspore.float32)).asnumpy()
+    for i in range(output.shape[1]):
+        assert np.all(output[:, i, :] == output_expect)
+    print("Dense layer wrapped successful")
+
+
+@pytest.mark.level0
+@pytest.mark.platform_x86_gpu_training
+@pytest.mark.env_onecard
+def test_time_distributed_dense_with_reshape_axis_not_first():
+    inputs = np.random.randint(0, 10, [32, 10])
+    dense = nn.Dense(10, 6)
+    output_expect = dense(Tensor(inputs, mindspore.float32)).asnumpy()
+    inputs = inputs.reshape([1, 32, 10]).repeat(6, axis=0)
+    time_distributed = TestTimeDistributed(dense, time_axis=0, reshape_with_axis=1)
+    output = time_distributed(Tensor(inputs, mindspore.float32)).asnumpy()
+    for i in range(output.shape[0]):
+        assert np.all(output[i, :] == output_expect)
+    print("Dense layer wrapped successful")
+
+
+@pytest.mark.level0
+@pytest.mark.platform_x86_gpu_training
+@pytest.mark.env_onecard
+def test_time_distributed_argmax():
+    inputs = np.random.randint(0, 10, [3, 4])
+    argmax = ops.Argmax(output_type=mindspore.int32, axis=1)
+    output_expect = argmax(Tensor(inputs, mindspore.float32)).asnumpy()
+    inputs = inputs.reshape([3, 1, 4]).repeat(6, axis=1)
+    time_distributed = TestTimeDistributed(argmax, time_axis=1, reshape_with_axis=0)
+    output = time_distributed(Tensor(inputs, mindspore.float32)).asnumpy()
+    for i in range(output.shape[1]):
+        assert np.all(output[:, i] == output_expect)
+    print("Argmax op wrapped successful")
+
+
+@pytest.mark.level0
+@pytest.mark.platform_x86_gpu_training
+@pytest.mark.env_onecard
+def test_time_distributed_flatten():
+    inputs = np.random.randint(0, 10, [3, 4, 5])
+    flatten = nn.Flatten()
+    output_expect = flatten(Tensor(inputs, mindspore.float32)).asnumpy()
+    inputs = inputs.reshape([3, 1, 4, 5]).repeat(6, axis=1)
+    time_distributed = TestTimeDistributed(flatten, time_axis=1, reshape_with_axis=0)
+    output = time_distributed(Tensor(inputs, mindspore.float32)).asnumpy()
+    for i in range(output.shape[1]):
+        assert np.all(output[:, i, :] == output_expect)
+    print("Flatten op wrapped successful")
+
+
+@pytest.mark.level0
+@pytest.mark.platform_x86_gpu_training
+@pytest.mark.env_onecard
+def test_time_distributed_conv2d_no_reshape_axis():
+    inputs = np.random.randint(0, 10, [32, 12, 10, 10])
+    conv2d = nn.Conv2d(12, 24, 4, has_bias=False, weight_init='normal')
+    output_expect = conv2d(Tensor(inputs, mindspore.float32)).asnumpy()
+    inputs = inputs.reshape([32, 1, 12, 10, 10]).repeat(6, axis=1)
+    time_distributed = TestTimeDistributed(conv2d, time_axis=1)
+    output = time_distributed(Tensor(inputs, mindspore.float32)).asnumpy()
+    for i in range(output.shape[1]):
+        assert np.all(output[:, i, :] == output_expect)
+    print("Conv2D layer with no reshape axis wrapped successful")
+
+
+@pytest.mark.level0
+@pytest.mark.platform_x86_gpu_training
+@pytest.mark.env_onecard
+def test_time_distributed_maxpool2d_no_reshape_axis():
+    inputs = np.random.randint(0, 10, [32, 12, 10, 10])
+    pool = nn.MaxPool2d(kernel_size=3, stride=1)
+    output_expect = pool(Tensor(inputs, mindspore.float32)).asnumpy()
+    inputs = inputs.reshape([32, 1, 12, 10, 10]).repeat(6, axis=1)
+    time_distributed = TestTimeDistributed(pool, time_axis=1)
+    output = time_distributed(Tensor(inputs, mindspore.float32)).asnumpy()
+    for i in range(output.shape[1]):
+        assert np.all(output[:, i, :] == output_expect)
+    print("MaxPooling2D layer with no reshape axis wrapped successful")
+
+
+@pytest.mark.level0
+@pytest.mark.platform_x86_gpu_training
+@pytest.mark.env_onecard
+def test_time_distributed_dense_no_reshape_axis():
+    inputs = np.random.randint(0, 10, [32, 10])
+    dense = nn.Dense(10, 6)
+    output_expect = dense(Tensor(inputs, mindspore.float32)).asnumpy()
+    inputs = inputs.reshape([32, 1, 10]).repeat(6, axis=1)
+    time_distributed = TestTimeDistributed(dense, time_axis=1)
+    output = time_distributed(Tensor(inputs, mindspore.float32)).asnumpy()
+    for i in range(output.shape[1]):
+        assert np.all(output[:, i, :] == output_expect)
+    print("Dense layer with no reshape axis wrapped successful")
+
+
+@pytest.mark.level0
+@pytest.mark.platform_x86_gpu_training
+@pytest.mark.env_onecard
+def test_time_distributed_argmax_no_reshape_axis():
+    inputs = np.random.randint(0, 10, [3, 4])
+    argmax = ops.Argmax(output_type=mindspore.int32, axis=1)
+    output_expect = argmax(Tensor(inputs, mindspore.float32)).asnumpy()
+    inputs = inputs.reshape([3, 1, 4]).repeat(6, axis=1)
+    time_distributed = TestTimeDistributed(argmax, time_axis=1)
+    output = time_distributed(Tensor(inputs, mindspore.float32)).asnumpy()
+    for i in range(output.shape[1]):
+        assert np.all(output[:, i] == output_expect)
+    print("Argmax op with no reshape axis wrapped successful")
+
+
+@pytest.mark.level0
+@pytest.mark.platform_x86_gpu_training
+@pytest.mark.env_onecard
+def test_time_distributed_flatten_no_reshape_axis():
+    inputs = np.random.randint(0, 10, [3, 4, 5])
+    flatten = nn.Flatten()
+    output_expect = flatten(Tensor(inputs, mindspore.float32)).asnumpy()
+    inputs = inputs.reshape([3, 1, 4, 5]).repeat(6, axis=1)
+    time_distributed = TestTimeDistributed(flatten, time_axis=1)
+    output = time_distributed(Tensor(inputs, mindspore.float32)).asnumpy()
+    for i in range(output.shape[1]):
+        assert np.all(output[:, i, :] == output_expect)
+    print("Flatten op with no reshape axis wrapped successful")